建立了一种基于民生视角的政府绩效评价贝叶斯网络模型。研究了运用极大熵模型,求解网络的参数分布问题;结合灰色系统的思想,给出了政府绩效评价中的不确定信息处理的方法;通过对网络结构的分析,研究了评价信息在传递过程中的衰减性;最后,通过贝叶斯网络推理,给出网络中各节点的定量评价,为基于民生视角的政府绩效评价提供了一种新的理论方法和工具。
This paper establishes a Bayesian network model to evaluate government performance from a livelyhood perspective. The parameter distribution problem is slovingby using Maximum Entropy Model. Then a method is proposed to deal with uncertain informations in the evaluation process according to grey system principles. The author also studies on the evaluation information decay in transfer process by analyzing network structure. Finally, the paper provides a new theory and tool for the government performance evaluation from a livelyhood perspective based on the quantitative evaluation of every network node throgh inference in Bayesian network.